Preparation

## Rows: 10886 Columns: 12
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## dbl  (11): season, holiday, workingday, weather, temp, atemp, humidity, wind...
## dttm  (1): datetime
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

Most frequently used days

It seems that there is no difference.

The relationship between different seasons and the number of car rentals

Average number of rentals per day in 24 hours per season

## `summarise()` has grouped output by 'season'. You can override using the
## `.groups` argument.

When temp is less than 35 Celsius, the correlation is positive. But if the temperature is too high, people don’t like to ride bikes.

How many people are renting a car per hour?

The relationship between temperature and rental volume

## `geom_smooth()` using method = 'gam' and formula = 'y ~ s(x, bs = "cs")'

The difference in count at 0-24

It is clear that around 9 a.m. and 6 p.m. is the peak period for rentals, which also coincides with people’s commuting times.

Correlation analysis

It can be seen temp is positively correlated with count, humidity and weather are negatively correlated with count.